Compelling evidence has demonstrated that the hippocampal formation and closely associated cortical regions in the medial temporal lobe critically participate in learning and memory. It remains unclear however whether the individual components of this system are all equally important for normal learning and memory and there is a marked lack of consensus concerning the specific mnemonic processing functions subserved by these structures. The proposed research will address these issues in a series of studies in the monkey (Macaca fascicularis). The principal hypothesis under investigation is that the hippocampus and the entorhinal, perirhinal and parahippocampal cortices mediate distinct and dissociable aspects of learning and memory. This account suggests the hippocampus selectively participates in storing representations of the relationships between items in memory, and supports the ability to use this representational organization flexibly in making inferences and generalizations from memory. The cortical components of the system are viewed as subserving a different function involving the maintenance of information over delays that exceed the span of immediate memory. To evaluate this proposal, new behavioral testing procedures will be developed for examining relational memory processes and the flexible expression of memory in monkeys. One set of tasks will be conducted in a manual test apparatus using procedures designed to assess paired associate learning, learning and memory for hierarchical relationships, and sequential order learning. Another set of tasks will utilize a large open field maze to provide the first systematic evaluation of learning and memory for spatial relations in the freely moving monkey. Probe tests will be incorporated in each procedure to confirm that performance is based on relational memory processes. Behavioral results from more standard learning and memory tasks (delayed non-matching-to-sample and retention of object discriminations) will provide baseline data for assessing the sensitivity and selectivity of relational memory tasks to hippocampal system damage. The proposed project will capitalize on this newly developed test battery to define the specific memory processing functions subserved by individual components of the hippocampal system. Separate groups of monkeys will be prepared with selective lesions of the fornix or the hippocampus (including the dentate gyrus and subicular,complex). or with combined damage to the perirhinal and parahippocampal cortices. Behavioral results from these groups will be compared with the effects of larger lesions involving both the hippocampal and cortical components of the system. The potential for further functional specialization within the hippocampal system will be evaluated in experiments examining the consequences of selective entorhinal, perirhinal, or parahippocampal cortical lesions. Taken together, these studies should lead to a more precise specification of the memory functions mediated by individual components of the hippocampal system, and should help bring into correspondence results from laboratory animal models and neuropsychological research on human amnesia.